Optimization of a Monitoring Network for Sulfur Dioxide

E. P. J. Boer, A. L. M. Dekkers, A. Stein
2002 Journal of Environmental Quality  
lation technique. This flexible method allows the characterization of trends by using simple local models. The In this study, we develop and apply a methodology to reduce an variance of estimates resulting from this method can be existing monitoring network to find an optimal configuration of a used to formulate a criterion for optimization. smaller network. We use a criterion based on locally weighted regression with two different weight functions. The methodology is applied The aim of this
more » ... dy is to further explore the possibilto the Dutch national SO 2 network and offers the possibility to include ities of locally weighted regression for the optimization different politically relevant options in the model by weight criteria. of a monitoring network (Mü ller, 1995; Fedorov et al., Because full enumeration of all monitoring networks is impossible, 1999). We show that it leads to a flexible criterion for a combinatorial search algorithm is applied to find a (sub)optimal adaptation of an existing monitoring network. This flexisolution. If different (optimal) monitoring networks result from differbility is expressed in the application of this interpolation ent criteria, then the best can be selected. technique and by incorporating different design criteria, which can easily include external (policy) constraints. The criterion is applied to a case study in the Nether-
doi:10.2134/jeq2002.1210 pmid:11841060 fatcat:5pf3fmz7dfgapju2ykxqsvd7ea